That's intriguing, I'd be curious to see a profile. Maybe for large
images it is actually worse.
One thing that might help would be to make `u` a shared variable and to
update it in `lap_and_err`, you may save a memory copy, but that may not
be a big deal.
On Mon, Oct 17, 2016, Bogdan Opanchuk wrote:
> Hi Pascal,
>
> Thanks for the suggestion. Paradoxically though, I get 4 times worse
> performance with nnet.conv2d: 21.3s vs 5.7s with the old nnet.conv.conv2d.
> The new function constructor that I have:
>
> from theano.tensor.nnet import conv2d
>
> ...
>
> def prepare_function_conv(dxd, dyd):
>
> flt = theano.shared(numpy.array([[[[0, dxd, 0], [dyd, 0, dyd], [0, dxd,
> 0]]]]))
> u = T.dmatrix('u')
>
> nx = u.shape[0]
> ny = u.shape[1]
>
> conv_res = conv2d(u.reshape((1, 1, nx, ny)), flt, border_mode='valid')
> conv_res = conv_res.reshape((nx - 2, ny - 2))
>
> u_new = T.set_subtensor(u[1:-1,1:-1], conv_res)
>
> v = u_new - u
> err = ((v**2).sum())**0.5 / u.size
>
> lap_and_err = theano.function([u], [u_new, err])
> return lap_and_err
>
>
>
> --
>
> ---
> You received this message because you are subscribed to the Google Groups
> "theano-users" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to [email protected].
> For more options, visit https://groups.google.com/d/optout.
--
Pascal
--
---
You received this message because you are subscribed to the Google Groups
"theano-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
For more options, visit https://groups.google.com/d/optout.
